mirror of
https://github.com/likelovewant/ollama-for-amd.git
synced 2025-12-23 23:18:26 +00:00
Update GGML to b6646 (#12245)
Notable EOLs with this change: - MacOS v12 and v13 are no longer supported (v14+ required) - AMD gfx900 and gfx906 are no longer supported
This commit is contained in:
@@ -13,10 +13,10 @@ checks.
|
||||
1 file changed, 18 insertions(+)
|
||||
|
||||
diff --git a/ggml/src/ggml-cuda/ggml-cuda.cu b/ggml/src/ggml-cuda/ggml-cuda.cu
|
||||
index 57eae461..c7f9dc3a 100644
|
||||
index ad389ece..e51c5035 100644
|
||||
--- a/ggml/src/ggml-cuda/ggml-cuda.cu
|
||||
+++ b/ggml/src/ggml-cuda/ggml-cuda.cu
|
||||
@@ -2671,12 +2671,24 @@ static bool check_node_graph_compatibility_and_refresh_copy_ops(ggml_backend_cud
|
||||
@@ -2686,14 +2686,26 @@ static bool check_node_graph_compatibility_and_refresh_copy_ops(ggml_backend_cud
|
||||
// Loop over nodes in GGML graph to obtain info needed for CUDA graph
|
||||
cuda_ctx->cuda_graph->cpy_dest_ptrs.clear();
|
||||
|
||||
@@ -36,12 +36,14 @@ index 57eae461..c7f9dc3a 100644
|
||||
const std::string ffn_moe_gate_bias_prefix = "ffn_moe_gate_biased";
|
||||
const std::string ffn_moe_up_bias_prefix = "ffn_moe_up_biased";
|
||||
const std::string ffn_moe_down_bias_prefix = "ffn_moe_down_biased";
|
||||
const std::string nemotron_h_block_out_prefix = "nemotron_h_block_out";
|
||||
const std::string mamba2_y_add_d_prefix = "mamba2_y_add_d";
|
||||
|
||||
+
|
||||
for (int i = 0; i < cgraph->n_nodes; i++) {
|
||||
ggml_tensor * node = cgraph->nodes[i];
|
||||
|
||||
@@ -2700,6 +2712,12 @@ static bool check_node_graph_compatibility_and_refresh_copy_ops(ggml_backend_cud
|
||||
@@ -2717,6 +2729,12 @@ static bool check_node_graph_compatibility_and_refresh_copy_ops(ggml_backend_cud
|
||||
|
||||
if (node->op == GGML_OP_ADD &&
|
||||
node->src[1] && node->src[1]->ne[1] > 1 &&
|
||||
|
||||
Reference in New Issue
Block a user